[论文解读] A Comparison of Image Denoising Methods

Zhaoming Kong, Fangxi Deng+4 more|arXiv (Cornell University)|Apr 18, 2023

Image and Signal Denoising Methods被引用 8

一句话总结

本文在合成与真实数据上比较了包括传统方法与深度神经网络在内的大范围图像去噪方法，并引入新的基准数据集。

ABSTRACT

The advancement of imaging devices and countless images generated everyday pose an increasingly high demand on image denoising, which still remains a challenging task in terms of both effectiveness and efficiency. To improve denoising quality, numerous denoising techniques and approaches have been proposed in the past decades, including different transforms, regularization terms, algebraic representations and especially advanced deep neural network (DNN) architectures. Despite their sophistication, many methods may fail to achieve desirable results for simultaneous noise removal and fine detail preservation. In this paper, to investigate the applicability of existing denoising techniques, we compare a variety of denoising methods on both synthetic and real-world datasets for different applications. We also introduce a new dataset for benchmarking, and the evaluations are performed from four different perspectives including quantitative metrics, visual effects, human ratings and computational cost. Our experiments demonstrate: (i) the effectiveness and efficiency of representative traditional denoisers for various denoising tasks, (ii) a simple matrix-based algorithm may be able to produce similar results compared with its tensor counterparts, and (iii) the notable achievements of DNN models, which exhibit impressive generalization ability and show state-of-the-art performance on various datasets. In spite of the progress in recent years, we discuss shortcomings and possible extensions of existing techniques. Datasets, code and results are made publicly available and will be continuously updated at https://github.com/ZhaomingKong/Denoising-Comparison.

研究动机与目标

• 评估传统去噪方法与基于DNN的方法在不同数据集与应用中的有效性。
• 为图像、视频、MSI/HSI和MRI去噪任务引入真实世界基准数据集。
• 使用定量指标、视觉效果、人类评级和计算成本来评估方法。
• 就去噪技术的泛化性与实际适用性提供观察。

提出的方法

• 将去噪方法分为传统与基于DNN的方法并进行分类。
• 使用分组-协同过滤-聚合框架描述传统去噪方法。
• 讨论NLSS先验与各种代数表示（矩阵、张量、基于SVD、低秩等）。
• 考察DNN的训练策略（有监督、自监督、无监督）及其对性能的影响。
• 引入新的真实世界数据集与多任务去噪基准。
• 在多模态数据的合成与真实世界实验中进行比较。

实验结果

研究问题

• RQ1 传统去噪方法与基于DNN的去噪方法在图像、视频、MSI/HSI与MRI去噪任务中的对比如何？
• RQ2 NLSS先验、变换与张量表示在去噪性能中的作用是什么？
• RQ3 预训练或数据集特定的DNN模型在新数据集与应用中的泛化能力如何？

主要发现

• 在若干真实世界数据集上，DNN方法显示出显著改进，但预训练模型在不同数据集之间的泛化性可能较差。
• 如BM3D族等传统去噪方法在多项去噪任务中仍具有竞争力。
• 改进的SVD（M-SVD）在相对于张量方法的竞争性结果。
• 某些DNN模型如FCCF、RVRT、FastDVDNet在真实世界的图像与视频去噪任务中取得了较强的结果。
• 如FastDVDNet、FloRNN、RVRT等视频去噪方法在效果与效率方面表现出色。